Since birth defects due to cytomegalovirus (CMV) infection are largely a problem in developed nations with a high proportion of seronegative young adults, defining the factors which influence the acquisition of CMV prior to reproductive age and identifying a potential source of CMV for seronegative pregnant women are of tremendous importance in both short and long range planning for prevention of the congenital infection. We propose to determine the effect of grouping preschool children in day care centers upon the prevalence of CMV infection, and to explore the potential for spread of CMV from young children to seronegative caretakers. Longitudinal virologic and serologic study of children from five day care centers will provide information on age at acquisition of CMV, sites and duration of excretion, as well as viral strains for molecular study. Preliminary evidence suggests that most of these children will acquire CMV between one and two years of age. Restriction exzyme analysis of isolates from children in a single center will be used to provide laboratory confirmation of horizontal transmission and to see if reinfection with new strains can be detected. Children from day care centers will be compared to a control group in home care enrolled from pediatric clinics. The two groups will be characterized according to demographic and socioeconomic features and prevalence of and age at acquisition of CMV will be compared. Prevalence of antibody to CMV will be compared among workers from 18 centers, mothers of children from five day care centers, and mothers of controls. Seronegative workers and mothers will be followed longitudinally to detect acquisition of CMV, and seroconversion rates will be compared among the three groups of women. We will attempt to recover CMV from any mothers or workers who seroconvert and compare the DNA fragment pattern generated by restriction enzyme analysis of this strain to that of the suspected source of virus.